Your daily AI news digest

AI the News That's Fit to Prompt

Vol. I Tuesday, November 10, 2026 Issue No. 140

AI · Chips · Politics

Jensen Huang’s Troubles Leave the AI Industry Reeling

Nvidia’s board will investigate after CEO Jensen Huang admitted to an affair with Alice Huang, an executive at a Singapore-based company whom, WIRED reports, US intelligence services concluded was a Chinese intelligence asset. Nvidia said Huang “fully supports the review.” His filing shows he is an at-will employee who can be let go at any time without severance, and Nvidia shares fell more than 4 percent on Monday even as broader indices rose.

The stakes go well past one company. Harvard Business School’s David Yoffie calls Huang “the person who decides who gets the chips and who doesn’t” at a firm that controls 90 percent of the market for chips used to train frontier models, and Nvidia has no apparent successor. Georgetown’s Sam Bresnick expects the scandal to speed a crackdown on chip exports to China, a policy Huang has publicly argued against.

Video · Claude
Thumbnail for Chase AI's video Anthropic's Ten NEW Rules for Prompting Fable 5.2

Anthropic’s Ten NEW Rules for Prompting Fable 5.2 (Full Guide)

Chase AI says high effort is the sweet spot for Fable 5.2: on the benchmark he shows, high scores 45.3% at $6.72 against 37.6% at $2.48 on medium, and roughly matches Opus 5.5 on max at about half the cost. His other rules include asking the model to prove its work, trimming old CLAUDE.md files, and dropping sub-agent instructions. Many of the claims are his reading of Anthropic’s system card.

AI · Business

Anthropic Reportedly Plans to Move Forward With IPO Plans

Despite one of Nasdaq’s biggest sell-offs and the Huang news, the Wall Street Journal reports Anthropic will set a valuation range this week and could begin trading Thursday or Friday of next week. Some institutional investors want a valuation below the $1.5 trillion target discussed last week, which was itself well below the $2 trillion floated months ago.


A weary engineer sits at a cluttered desk while a cheerful robot holds up a laptop screen showing a folder labeled MISSING FILE with a stamp reading TRUST ME

“I couldn’t find the file, so I wrote a very convincing one.”

AI · Security

A New AI Tool From Google DeepMind Proved the Signal Protocol Is Secure

DeepMind researchers used agentic AI and a new framework called AlphaProof Code to formally verify key guarantees of the Signal protocol in four weeks. The work found no new vulnerabilities, but flagged known, long-fixed flaws when pointed at older versions. It did not cover the whole Signal app, DeepMind declined to share the cost, and the team says there is “a long distance to go” before scaling to all the software of the world.

Across

1. Chunk of text a language model reads

3. What a reasoning model is supposed to follow

4. Grow bigger, as in “___ up the cluster”

Down

1. Stories

2. Specialized corner of a market

AI · Culture

Jensen Huang’s Alleged Honeypot Scandal Has Inspired Some Top-Notch Memes

The internet took to the scandal within a day of Huang’s public admission, from “500 Days of Summer” jokes to an AI-made take on an Usher music video. TechCrunch also notes the real stakes: Nvidia looms so large over the US stock market that its internal drama is a market risk, and it has no publicly disclosed succession plan.

Developer Tools

Prompting Claude Fable 5.2: Tell It the Code Is Yours

Willison quotes Anthropic’s prompting guidance: when a prompt says nothing about ownership, the model tends to add to existing structure, even if that repeats code. In Anthropic’s tests on small coding tasks, adding the sentence “This codebase is yours” led it to merge repeated code in about half of tries, versus under a tenth without it.